Yuquan Gan
ABSTRACT
In the field of hyperspectral anomaly detection, autoencoder (AE) have become a hot research topic due to their unsupervised characteristics and powerful feature extraction capability. However, autoencoders do not keep the spatial structure information of the original data well during the training process, and is affected by anomalies, resulting in poor detection performance. To address these problems, a hyperspectral anomaly detection method based on autoencoders with superpixel manifold constraints is proposed. Firstly, superpixel segmentation technique is used to obtain the superpixels of the hyperspectral image, and then the manifold learning method is used to learn the embedded manifold that based on the superpixels. Secondly, the learned manifold constraints are embedded in the autoencoder to learn the potential representation, which can maintain the consistency of the local spatial and geometric structure of the hyperspectral images (HSI). Finally, anomalies are detected by computing reconstruction errors of the autoencoder. Extensive experiments are conducted on three datasets, and the experimental results show that the proposed method has better detection performance than other hyperspectral anomaly detectors.
- Kang Sun, Xiurui Geng, Luyan Ji and Yun Lu. 2014. A new band selection method for hyperspectral image based on data quality. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 7.6 (2014): 2697-2703.Google Scholar
- Yuan, Yuan, Dandan Ma and Qi Wang. 2015. Hyperspectral anomaly detection by graph pixel selection. IEEE transactions on cybernetics 46.12 (2015): 3123-3134.Google Scholar
- Xiaoqiang Lu, Wuxia Zhang and Xuelong Li. 2017. A hybrid sparsity and distance-based discrimination detector for hyperspectral images. IEEE Transactions on Geoscience and Remote Sensing 56.3 (2017): 1704-1717.Google Scholar
- Telmo Adão, Jonáš Hruška, Luís Pádua, José Bessa, Emanuel Peres, Raul Morais and Joaquim J. Sousa. 2017. Hyperspectral imaging: A review on UAV-based sensors, data processing and applications for agriculture and forestry. Remote sensing 9.11 (2017): 1110.Google Scholar
- Shangzhen Song, Huixin Zhou, Yixin Yang and Jiangluqi Song. 2019 Hyperspectral anomaly detection via convolutional neural network and low rank with density-based clustering. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 12.9 (2019): 3637-3649.Google Scholar
- Wei Li, Guodong Wu and Qian Du. 2017. Transferred deep learning for anomaly detection in hyperspectral imagery. IEEE Geoscience and Remote Sensing Letters 14.5 (2017): 597-601.Google Scholar
- Shangzhen Song, Huixin Zhou, Yixin Yang and Jiangluqi Song. 2019. Hyperspectral anomaly detection via convolutional neural network and low rank with density-based clustering. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 12.9 (2019): 3637-3649.Google Scholar
- Chunhui Zhao, Xueyuan Li and Haifeng Zhu. 2017. Hyperspectral anomaly detection based on stacked denoising autoencoders. Journal of Applied Remote Sensing 11.4 (2017): 042605.Google Scholar
- Ning Ma, Yu Peng, Shaojun Wang and Philip H.W. Leong. 2018. An unsupervised deep hyperspectral anomaly detector. Sensors 18.3 (2018): 693.Google Scholar
- Fei Li, Xiuwei Zhang, Lei Zhang, Dongmei Jiang and Yanning Zhang. 2018. Exploiting structured sparsity for hyperspectral anomaly detection. IEEE Transactions on Geoscience and Remote Sensing 56.7 (2018): 4050-4064.Google Scholar
- Chunhui Zhao and Lili Zhang. 2018. Spectral-spatial stacked autoencoders based on low-rank and sparse matrix decomposition for hyperspectral anomaly detection. Infrared Physics & Technology 92 (2018): 166-176.Google ScholarCross Ref
- Xiaoqiang Lu, Wuxia Zhang and Ju Huang. 2019. Exploiting embedding manifold of autoencoders for hyperspectral anomaly detection. IEEE Transactions on Geoscience and Remote Sensing 58.3 (2019): 1527-1537.Google Scholar
- Jie Lei, Weiying Xie, Jian Yang, Yunsong Li and Chein-I Chang. 2019. Spectral–spatial feature extraction for hyperspectral anomaly detection. IEEE Transactions on Geoscience and Remote Sensing 57.10 (2019): 8131-8143.Google Scholar
- Chris Ding, Ding Zhou, Xiaofeng He and Hongyuan Zha.2006. R1-pca: rotational invariant l 1-norm principal component analysis for robust subspace factorization. Proceedings of the 23rd international conference on Machine learning. 2006.Google Scholar
- Charles M Bachmann, Thomas L. Ainsworth and Robert A. Fusina. 2005. Exploiting manifold geometry in hyperspectral imagery. IEEE transactions on Geoscience and Remote Sensing 43.3 (2005): 441-454.Google Scholar
- Jun He, Lei Zhang, Qing Wang and Zigang Li. 2009. Using diffusion geometric coordinates for hyperspectral imagery representation. IEEE Geoscience and Remote Sensing Letters 6.4 (2009): 767-771.Google Scholar
- Li Ma, Melba M. Crawford and Jinwen Tian. 2010. Local manifold learning-based k-nearest-neighbor for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing 48.11 (2010): 4099-4109.Google Scholar
- Li Ma, Melba M. Crawford, Xiaoquan Yang and Yan Guo. 2014. Local-manifold-learning-based graph construction for semisupervised hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing 53.5 (2014): 2832-2844.Google Scholar
- Hong-Bing Huang, Hong Huo and Tao Fang. 2013. Hierarchical manifold learning with applications to supervised classification for high-resolution remotely sensed images. IEEE Transactions on Geoscience and Remote Sensing 52.3 (2013): 1677-1692.Google Scholar
- Yuan Yan Tang, Haoliang Yuan and Luoqing Li. 2014. Manifold-based sparse representation for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing 52.12 (2014): 7606-7618.Google Scholar
- Li Ma, Melba M. Crawford and Jinwen Tian. 2010. Anomaly detection for hyperspectral images based on robust locally linear embedding. Journal of Infrared, Millimeter, and Terahertz Waves 31.6 (2010): 753-762.Google Scholar
- Lefei Zhang, Liangpei Zhang, Dacheng Tao and Xin Huang. 2013. Sparse transfer manifold embedding for hyperspectral target detection. IEEE Transactions on Geoscience and Remote Sensing 52.2 (2013): 1030-1043.Google Scholar
- Xiaofeng Ren and Jitendra Malik. 2003. Learning a classification model for segmentation. Computer Vision, IEEE International Conference on. Vol. 2. IEEE Computer Society, 2003.Google Scholar
- Marcus S. Stefanou and John P. Kerekes. 2009. A method for assessing spectral image utility. IEEE Transactions on Geoscience and Remote Sensing 47.6 (2009): 1698-1706.Google Scholar
- Zhe Wang and Ryan Martin. 2020. Model-free posterior inference on the area under the receiver operating characteristic curve. Journal of Statistical Planning and Inference 209 (2020): 174-186.Google ScholarCross Ref
- Peter A. Flach, José Hernández-Orallo and Cèsar Ferri Ramirez. 2011. A coherent interpretation of AUC as a measure of aggregated classification performance. ICML. 2011.Google Scholar
Index Terms
- Hyperspectral Anomaly Detection based on Autoencoder using Superpixel Manifold Constraint
Recommendations
Maximized subspace model for hyperspectral anomaly detection
An important application in remote sensing using hyperspectral imaging system is the detection of anomalies in a large background in real-time. A basic anomaly detector for hyperspectral imagery that performs reasonaly well is the RX detector. In ...
Complementary Set Variational Autoencoder for Supervised Anomaly Detection
2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)Anomalies have broad patterns corresponding to their causes. In industry, anomalies are typically observed as equipment failures. Anomaly detection aims to detect such failures as anomalies. Although this is usually a binary classification task, the ...
Improved autoencoder for unsupervised anomaly detection
AbstractDeep autoencoder‐based methods are the majority of deep anomaly detection. An autoencoder learning on training data is assumed to produce higher reconstruction error for the anomalous samples than the normal samples and thus can distinguish ...
Comments